Empowering the Retina Patient with Use-at-Home Technology

New Technology Introduction

Sequence of Authors (1, 2, ..) / Name / Email ID / Phone No./ Fax No. / Designation / Affiliation
Corresponding Author / 1 / Christopher D. Hekimian, Dr. Sc. SE, MSEE / / 301 520-1575 / CEO and Chief Scientist / dxdt Engineering and Research

1

ABSTRACT

The paper Amsler grid has been in use consistently by retina patients since 1945. The paper Amsler grid is of limited utility based on difficulty with identifying and tracking retinal aberrations consistently, objectively and accurately. Now, with computers in so many homes, it is possible to capture all of the benefits of an Amsler grid while adding many more features useful to the patient and to the physician. The objective of our development initiative, carried out in consult with the National Eye Institute, was to develop a computerized replacement for the Amsler grid. The system, the Quantitative Retina Test Grid (QRTG)is used by patients at home tomeasure and track metrics associated with warp distortion and blind area in the field of vision. The new tool enables the patient to collect data over time relevant to meaningful trends in retina health.

Regular use of the tool allows physicians to develop prognoses that would have otherwise taken multiple visits to a clinic over the course of many months. The use of the QRTG for large scale clinical research could hold significant benefits in terms of diversity in metrics and order of magnitude increases in sample size.

Keywords: Amsler Grid, Quantitative Retina Test Grid, Macular Degeneration, dxdt, Hekimian

Abbreviations: QRTG: Quantitative Retina Test Grid; NEI: National Eye Institute; NIH: National Institutes of Health

1

  1. INTRODUCTION

This paper introduces a concept for a quantitative retina test grid that can be used by retina patients at home, in order to monitor retina health over time. The system is in essence a computerized replacement for a paper Amsler grid. Shortly after being diagnosed with Stargardt’s disease, the author collaborated with Dr. Wadih Zein and Dr. Catherine Meyerle of the National Eye Institute of the National Institutes of Health on a concept for a quantitative retina test grid. The result of that collaboration was a low-cost system that has the potential to replace a paper Amsler grid. The new system, described in this paper utilizes features available on a computer to identify scotomae, and to measure total blind area and the degree of warp distortion perceived in the grid lines for each eye and to track these metrics over time. We believe that the tool represents an improvement in terms of at-home retina health testing and retina health tracking. The tool has been used for developing retina health prognoses and dxdt engineering and research has developed a trend analysis workbook for that purpose and to test for the efficacy of interventions. The tool was introduced in 2012 and has been supplemented with many new features. Dxdt Engineering and Research is actively looking for research partners that may be interested in studying various aspect of the Quantitative Retina Test Grid (QRTG).

1.1 Background

Paper Amsler grids have been in widespread use by ophthalmologists and their patients since 1945. Sometimes ophthalmologists will instruct their patients to use a pencil to mark areas of the Amsler grid that are affected by scotomae. They may also instruct the patient to indicate which lines in the Amsler grid appear bent or wavy. The ophthalmologist can apply their own measurements to the Amsler grid and track the metrics over time, looking for dramatic changes indicative of a worsening condition and a need for intervention. More often, the paper Amsler grid is provided to patients that are simply instructed to use the grid once a month and call the doctor if the appearance of the grid changes significantly.

1.2 Limitations of the Paper Amsler Approach

The Author finds the following problems with the paper Amsler approach:

a)If one wants to collect metrics from a paper Amsler grid, it is not a trivial matter to do so. Blind areas can be calculated crudely by counting up affected grid squares. Measuring the distortion of the grid lines would take some other means and would presumably be less straightforward. Therefore, tracking retina health over time using a paper Amsler grid is labor intensive and problematic.

b)It is not likely that retina patients that use the grid once a month will be able to remember what the previous months grid looked like with much fidelity. Hence, small or gradual changes in retina health will not be noticed. The monthly process is likely to be perceived as futile to many retina patients (like it did to the Author). Many patients will probably not bother to use the paper grid.

c)Even when a paper Amsler grid is used regularly as prescribed, it is hard for the patient to feel that they are doing anything worthwhile to preserve their eyesight by looking into the grid on a regular basis. The patient does not feel like they are tracking their eye health or learning anything new about their retina health and long or short term prognoses. This can contribute to a sense of helplessness and despair on behalf of the patient.

Another problem that was experienced by the Authorwas that having been diagnosed with a degenerative retina disease, no ophthalmologists were able or willing to offer a prognosis about the rate at which his eyesight was degrading. They simply did not have the data available to make such a prognosis. To get sufficient data, multiple visits to the clinic over the course of several months would likely be required. Insurance companies are not known to approve of so many visits for the purpose of issuing detailed prognoses. Consequently, the patient is left wondering how long his or her vision will support their current lifestyle.

1.3Objective of Development Effort

It was the objective of the researchers to address problems a) through c) with the development of the QRTG. We feel that the tool also solves the prognosis problem and has merit as a research instrument for large scale virtual and in –person clinical trials. The following sections describes several features of the QRTG that are intended to empower the retina patient with a tool that collects clinically meaningful data and also is supportive of personal safety and improved quality of life.

1.4Summary of Development Effort

Our researchers work in proximity to the National Institutes of Health (NIH) in Bethesda, MD. Upon being diagnosed with Stargardt’s disease, our Chief Scientist volunteered for studies involving Stargardt’s at the National Eye Institute (NEI) at NIH. It was during the course of these interactions that the subject of a QRTG was broached. The doctors at NEI encouraged the development effort and offered their services in helping to refine the methods that were eventually incorporated into the prototype system. After initial market introduction, the comments from a many more ophthalmologists led to a far more advanced system that is described in this paper.

  1. EXPERIMENTAL

2.1 The Prototype System

Using a personal computer is a natural progression from using a paper Amsler grid. The prototype system allowed for warp distortion measurements using a 10 x 10 grid. The number of grid lines was reduced from a standard Amsler grid because it was time consuming and cumbersome for a patient to do grid distortion measurements using more than about 100 grid lines. Both the prototype system and the current one measure warp distortion by allowing the patient to drag and distort the Amsler grid lines such that they compensate for the distortion perceived by the patient and ultimately appear straight to the afflicted eye. A single metric is calculated based on the average amount of compensation that was applied to the grid to make it appear straight.

2.2 Identifying Blind Spots (Scotomae)

The computer provides an ideal platform for identifying scotomae. They were readily identified using the prototype system either by observing which parts of the grid dropped out when one eye was closed, or by “wiggling” the mouse pointer and noting when it could not be seen. Mouse click and drag operations allowed for arbitrarily shaped areas to be drawn on the grid corresponding to scotomae.

2.3Maintaining Data Consistency

The grid was made of adjustable size and alternate color schemes were provided. It became evident that safeguards would have to be incorporated so that collected data for a given patient profile (data set) would be collected at a common eye to screen distance and with identical starting grids. The patient profiles would accumulate the patient measurements of blind area and average warp distortion for each eye as well as the images of the grids that the patients produced.

2.4Data Reporting

The prototype system also comprised a reporting screen which shows the tabulated measurement results for the blind area and warp distortion for each eye by date. Autoscaling line plots are also provided that show the trends in eye health metrics directly. A feature for easy export of the data tables was also incorporated. This prototype system was the basis for the first version of the system that was brought to market. It was provided pre-installed on its own USB drive. The system requires a computer that can run a Microsoft Windows operating system. Figure 1 shows the reporting screen with four line plots graphically depicting the two metrics for each eye. At the bottom of the form are the tabulated values for each eye.

Figure 1. Quantitative Retina Test Grid Reporting Screen

2.5Validation of Need for a QRTG

The early system was introduced at the 2012 American Academy of Ophthalmology and Asian Pacific Academy of Ophthalmology Joint Meeting held in Chicago, IL in November 2012. The system was received enthusiastically and discussions that the Author had with dozens of Ophthalmologists led to some notable improvements. No comparable capabilities were to be found at either the AAO-2012 or AAO-2013 international events. In particular, there is no capability for patients at home to collect and store retina health metrics associated with warp distortion and total blind area and communicate results by e-mail with physicians. Furthermore, no other systems provide the means to track retina health metrics or to collect, store, visualize and slew test images automatically by computer.

  1. RESULTS AND DISCUSSION

3.1 Improvements in Accessibility

The basic backbone and functionality of the prototype system endures in the current system. Figure 2shows the basic black on white test grid. Many changes were made to better accommodate individuals with limited vision. Included among them are two additional ways for identifying scotomae.

Figure 2. Basic Black on White Retina Test Grid Showing Marked Scotomae Areas and Grid Warpage.

3.2 Identifying Scotomae

The new system allows for identification of scotomae against a static background of small blocks in a checkerboard pattern. Areas of the retina afflicted by scotomae cannot discern the detail of the block pattern. Marking the scotomae involves drawing an outline around the boundary of where the detail in the grid is lost. See Figure 3.

Figure 3. Quantitative Retina Test Grid with Checkered Background

Another option included in the current system is to use a dynamic noise background. This option fills the background of the screen with rapidly moving “snow” like one would see on a television set with no signal. Areas of the retina afflicted by scotomae cannot see the motion of the snow. Marking scotomae in this way involves drawing an outline around the boundary of where the motion of the noise meets the visually “quiet” parts of the grid. Figure 4depicts the Quantitative Retina Test Grid with the dynamic noise background and red grid lines.

Figure 4. Quantitative Retina Test Grid with Dynamic Noise Background and Red Grid Lines

3.3 Off-Macula Visualization Trainer

Given that the system is capable of identifying areas of the central retina that are either healthy or unhealthy, it seemed reasonable to provide a simple tool that could be used by some patients to train themselves on visualizing using the healthy portions of their retinae. This system is simply a text box that can be arbitrarily sized and positioned anywhere relative to a central point of focus (like the center of an Amsler grid). Arbitrary text is loaded into the text box and the font size is made sufficiently large that it can be read off-macula. As the patient trains themself to read using the chosen part of their retina, the font size is incrementally reduced. The objective is to help the patient read normal print off-macula. See Figure 5.

Figure 5. Off-Macula Visualization Trainer Provided with the Retina Test Grid

3.4 Time-Lapse or Slewable Test Images

Due to a built in image file naming system, the grid images that the patient produces are easily slewed over time using slide show features that are built into Microsoft Windows operating systems.

3.5 Binocular Vision Analyzer

Since a given test session will likely involve testing each eye, the data existed to support binocular vision analysis. A feature has been included that allows any two grid images of a common size to be binocularly analyzed. A grid will show which areas of binocular central vision are blind to both eyes, which areas are blind to just one eye, and which areas are sighted to both eyes. This feature is included to support the day-to-day safety for the patient. Figure 6 shows the binocular vision analysis screen.

Figure 6. Binocular Vision Analysis Tool

  1. CONCLUSION

The current system supports retina health measurement, retina health tracking, reporting, safety and vision health training.

4.1 Need for Clinical Trials

The developers of the QRTG system are motivated to support all aspects of research associated with the use of the grid. In particular, we believe that it is necessary to conduct trials to quantify the typical variance associated with a patient’s day-to-day fidelity with which he/she can use the tool to accurately depict the same baseline retina condition. We know that besides measuring retina condition, the tool also measures how well the tool is used on a given day. This represents a validity issue that would have to be considered if the results were to be used to produce or confirm clinical findings.

4.2 Associated Analysis Support

Dxdt Engineering and Research has produced a trend analysis worksheet that can be used in conjunction with the retina test grid in order to test hypotheses on the slopes of regression lines, the effects of interventions, and on randomness of upticks and downticks in the data. The worksheet can be obtained by download at no charge from

4.3 Non-clinical Applications

For non-clinical application (i.e., use at home by patients)we anticipate the tool would be effective in keeping a retina patient or patients at risk of retinopathy meaningful engaged in a retina health monitoring regimen that would be more effective than the use of a paper Amsler grid. Furthermore, the built in features will go far to answer important questions for the patient outside of the clinical setting.

4.4 Clinical Applications

Subsequent to AAO-2013, several sales were made to ophthalmologists representing clinics from developing nations. This is taken as indicative of interest for the system to be used in a clinical setting. Clinic sized systems are available and a Spanish language version of the system is expected to be available in early 2014.

4.5 Comparison with Existing Products

While there are iPhone and tablet applications that replicate a paper Amsler grid, none have the capability to collect and store retina health metrics associated with warp distortion and total blind area. These systems are also subject to error based on inconsistencies in grid size and pupil to grid distance. The systems also lack the capability to store, display and communicate results by e-mail with physicians. The systems do not include off-macula visualization trainers or a means to show the patient their binocular vision risk areas. Criticisms of the system have been largely based on the assessment that elderly patients would not likely to be interested in using a computer to monitor their retina health at home.

4.6 Research Highlights

Highlights of the QRTG research and development efforts include the interest shown by the ophthalmological community demonstrated through the many comments and suggestions they have offered on how to improve and exploit the system. It has been encouraging to have been contacted by a team of researchers interested in using the system to support clinical research. Also, several systems have been sold to clinics in developing countries suggesting that the system will or is being used for clinical support.

4.7 Research Limitations

The research and development effort leading to the QRTG system described in this paper falls short of a formal validation of the QRTG approach, i.e., the notion of applying clinical purpose to data collected at home by patients, or the technical instantiation of the system itself. The developers of the QRTG simply offer the system as a highly functional replacement for the paper Amsler grid. Research and clinical use of the system should be preceded by or associated with rigorous validation of the approach. The developers are interested in supporting any such validation efforts.

4.8 Recommendations